Metastasis is the most lethal feature of human cancers, but the molecular processes that govern metastatic spread are not well understood. One of the major obstacles in this crucial area of cancer research is the lack of faithful genetic de novo cancer model that results in regional and distant metastases. Xenograft models based on cancer cell lines cannot fully recapitulate the intricate interplay in the complex microenvironment of stromal and cancer cells. In addition, the necessity of using immuno-compromised mice for xenograft transplants precludes elucidation of the likely critical interaction between the primary cancer and the immune system in cancer progression. The Sharpless and Wong laboratories have recently generated de novo mouse lung cancer and melanoma models driven by activated oncogenic Kras and concurrent tumor suppressor Lkb1 loss that has a >60% penetrance of regional and distant metastases. We demonstrated that loss of Lkb1 function is crucial in the promotion of cancer invasion and metastasis and dissected the signaling pathways that are crucial to this process. We have now further improved this genetic model by incorporating a p53 mutant allele into the Lkb1/Kras mutant mice. Compound mutant mice with Kras driven lung cancer or melanoma that have concurrent Lkb1 and p53 loss have distant hematogenous metastases (100% penetrance), thus demonstrating distinct and separate tumor suppressor functions for Lkb1 and p53. Distant organs harboring metastases include lymph nodes, spleen, kidney, liver, bone and brain, recapitulating the full spectrum of metastatic sites observed in association with human lung cancers and melanoma. With this proposal, we wish to further characterize and refine these metastatic cancer models and incorporate in vivo non-invasive imaging markers (luciferase and enhanced green fluorescent protein) into these compound mutant mice to permit tracking of the location of micro- metastasis non-invasively using the latest in vivo non-invasive imaging techniques. These models will then be used to test targeted therapeutics that can kill these cancers and inhibit metastatic spread. With collaborators in the UNC Center for Integrative Chemical Biology and Drug Discovery, we will also develop small molecule tool compounds that will enhance metastatic behavior by inhibiting Lkb1 activity. These studies will yield important insights into primary and acquired resistance of these cancers to the various treatments to help facilitate better treatment strategies. PUBLIC HEALTH RELEVANCE: Metastasis is the most lethal feature of cancer, but is poorly understood. Using faithful, genetically engineered models of lung cancer and melanoma, we will develop effective anti-metastasis therapies for in vivo use.